A R C H I V E S O F M E T A L L U R G Y A N D M A T E R I A L S

Volume 58 2013 Issue 4

DOI: 10.2478/amm-2013-0162

A. SZUDARSKA∗ , D. GURYNIUK∗ , T. MIZERSKI∗ , M. SZAFRAN∗

GELCASTING OF ALUMINA WITH APPLICATION OF NEW MONOMER SYNTHESIZED FROM XYLITOL

ODLEWANIE ŻELOWE TLENKU GLINU Z ZASTOSOWANIEM NOWEGO MONOMERU ZSYNTEZOWANEGO Z KSYLITOLU

A new monomer for application in the gelcasting process has been designed and synthesized, following the green chemistry
trend to use non toxic and renewable compounds. Xylitol was the main substrate for the synthesis that included 3 stages. The
final product, named 1-O-acryloylxylitol, was tested regarding its effectiveness as the processing agent for fabrication of
alumina elements by gelcasting method. Rheological properties of the ceramic slurries were studied as well as the properties
of shaped elements before and after sintering. The results are compared to those obtained by applying commercial monomer
2-hydroxyethyl acrylate.
Keywords: alumina, gelcasting, low-toxic monomers, xylitol

Podążając za trendem „zielonej chemii”, zakładającym m.in. wykorzystywanie nietoksycznych i odnawialnych surowców
w technologii, zaprojektowano i zsyntezowano nowy monomer przeznaczony do zastosowania w procesie odlewania żelowego.
Wyjściowym substratem trójetapowej syntezy był ksylitol, a finalnym produktem 1-O-akryloiloksylitol. Otrzymana pochodna
została sprawdzona pod względem efektywności polimeryzacji in situ w ceramicznej masie lejnej na bazie tlenku glinu. Zbadano
jej wpływ na właściwości reologiczne stosowanych zawiesin. Odlano serie kształtek, dla których wyznaczono podstawowe
parametry w stanie surowym oraz po spiekaniu. Wyniki porównano stosując równolegle komercyjnie dostępny monomer
akrylan 2-hydroksyetylu.

1. Introduction of industrial application, because the most effective systems

are highly toxic at the same time (eg. acrylamide as monomer
together with N,N’-methylenebisacrylamide as cross-linker).
The gelcasting technique has been researched for the last
That is one of the main reasons why the process has not been
two decades due to very attractive possibilities of inexpensive
industrialized yet for the large scale production and is often
forming of advanced ceramics of complex shape. The idea
applied only in the laboratories.
of the process is to carry out a controlled polymerization in
Monosaccharadies and their acryloyl derivatives have
situ in the ceramic suspension and create a macromolecular
been recently found very promising as a group of compounds
network that holds the ceramic particles together and ensures
and effective processing agents [5-8]. It was reported that the
high mechanical strength of the green bodies before sinter-
creation of a very strong macromolecular network is possible
ing [1-3]. Due to that, the machining of the gelcast is not
without the external cross-linking agents due to the presence
required and the method can be classified as ‘near-net-shape’
of many hydroxyl groups in their molecules and hydrogen
techniques. Originally, the term ‘gelcasting’ referred to the
bonds between them. Moreover, it appeared that they also
process where the polymerization of acrylic monomer together
can be dispersants for alumina nano- and submicropowders.
with a cross-linking agent was initiated by chemical reaction.
Monosaccharides are polyhydroxycompunds, as well as poli-
The constant development of the gelcasting spread the idea to
hydroxyalcohols, but no monomer based on polihydroxyalco-
consolidate the suspension by the chemical or physical gela-
hol has yet been synthesized and tested. The only near example
tion of many different compounds, initiated chemically, photo-
could be glycerol monoacrylate, that has two hydroxyl groups
chemically or thermally. Recent researches concerning the gel-
in the molecule and was found very effective as monomer in
casting, mainly focus on three aspects: (1) the development of
the gelcasting process [8]. That is the reason why xylitol, a
low/non-toxic gelating systems that don’t require thermal initi-
polyhydroxyalcohol, was chosen for the present research. It
ation, (2) the methods that allow one to reduce the possibility
is a non-toxic, renewable, water-soluble and inexpensive sub-
of causing some defects, especially during forming products of
stance. The paper presents its synthesis, as well as application
a large shape, (3) combining it with the other techniques and
in fabrication of alumina elements by gelcasting method. The
applying it in new areas (eg. fabrication of porous ceramics)
influence of the monomer content in the suspension is dis-
[4]. The first one is the most crucial regarding the possibility
∗
WARSAW UNIVERSITY OF TECHNOLOGY, FACULTY OF CHEMISTRY, 3 NOAKOWSKIEGO STR., 00-664 WARSZAWA, POLAND
1300

cussed and the main results concerning the properties of the
suspensions and gelcast elements are compared to those ob-
tained with the application of commercial low-toxic monomer
2-hydroksyethyl acrylate.
2. Experimental
2.1. The synthesis of the new compound
The synthesis route of 1-O-acyrloylxylitol (Akr-X) is
shown in Fig. 1. The aim of the synthesis, was the selective
substitution of acryloyl group instead of hydroxyl group to the
xylitol molecule. It was carried out in three stages, analogous
to the synthesis route reported for 3-O-acryloyl-D-glucose [5].
In the first stage (Fig. 1a) was aimed at blocking four from the
five hydroxyl groups in xylitol molecule by the reaction with
acetone in acidic medium. The second stage was the esterifi-
cation of the only remaining hydroxyl group, done by acryloyl
chloride in the presence of amine. It allows to introduce the
functional group with the double bond to the molecule. The
third stage (Fig. 1c) was focused on acidic hydrolysis that
unlocks the hydroxyl groups. The final product is ready to
use as an aqueous solution (20%) or as a pure monomer after
evaporating the solvent.
Fig. 1. The synthesis of 1-O-acryloylxylitol (Akr-X)
2.2. Materials and other components of the slurries
mean particle size D50 = 0.15 µm and specific surface area
SBET = 14.5 m2 /g. Diammonium hydrocitrate, DAC (POCh,
Poland) and citric acid, CA (Sigma) were used as disper-
sants. N,N,N’,N’-tetramethylethylene-diamine, TEMED (Flu-
ka) played the role of activator and ammonium persulfate,
AP (Aldrich) used in the form of 1wt % aqueous solution,
was the initiator of polymerization. Two monomers were
applied, newly synthesised Akr-X and commercially avail-
able 2-hydroksyethyl acrylate, HEA (Fluka) accompanied with
a cross-linking agent, N,N’-methylenebisacrylamide, MBAM
(Fluka). The molecules of that gelating system is shown in
Fig. 2.
Fig. 2. Chemical formula of a) 2-hydroxyethyl acrylate (HEA) and
b) N,N’-methylenebisacrylamide (MBAM)
2.3. Shaping procedure
Firstly, ceramic slurries were prepared at room tempera-
ture in the containers made of alumina. The dispersants, the
monomer and activator were dissolved in distilled water and
the ceramic powder was added at last. The contents of all
additives are listed in Table 1. The slurries were mixed in
a planetary ball mill (PM100, Retch) for 60 minutes with
a speed of 300 rpm. Next, the slurries were degassed un-
der reduced pressure in a vacuum desiccator supplied with a
magnetic stirrer. Just before casting the slurries into the round
plastic moulds (ϕ20), the initiator was added and they were
mixed for 3 minutes by magnetic stirrer. The access of the
fresh air was cut off by placing the mould filled with slurries
under cover for 6h to complete the gelation process. After the
gelled bodies were obtained and removed from the moulds
they were dried in air at room temperature for 24h, then at
50◦ C for 24h. Just before sintering they were dried at 105◦ C
for 1h. The specimens were sintered at 1500◦ C for 2h.

The ceramic powder used in the research was a high

purity α-Al2 O3 TM-DAR (Taimei Chemicals Ltd., Japan) of

TABLE 1
The composition of the ceramic suspensions prepared for the gelcasting

Slurry Al2 O3 DAC + CA Akr-X/ HEA TEMED MBAM APS

powder dispersant monomer activator cross-linker initiator
(denotation) [vol %] [wt %+ wt %]∗ [wt %]∗ [wt %]∗∗ [wt %]∗∗ [wt %]∗∗
TX3 50 0.30 + 0.10 3.0 (Akr-X) 2.0 - 0.3
TX5 50 0.30 + 0.10 5.0 (Akr-X) 2.0 - 0.3
THA3 53 0.14 + 0.10 3.0 (HEA) 2.0 2.0 0.1
THA5 53 0.14 + 0.10 5.0 (HEA) 2.0 2.0 0.1
∗ ∗∗
amount with respect to the powder content, amount with respect to the monomer content

2.4. Characterization techniques
The viscosities of the suspensions after mixing and de-
gassing, but before initializing of the polymerization, were
measured by using Brookfield DV+II- Pro rheometer (Brook-
field Engineering Laboratories Inc. Massachusetts, USA) as
a function of shear rate. The idle time was calculated on the
basis of viscosity measurements in time at a constant shear
rate 0.1s−1 , 3 minutes after addition of the initiator to the sus-
pension. The densities of green and sintered specimens were
measured by the Archimedes’ method in kerosene. The tensile
strength of the green bodies was determined by the ‘Brazilian
test’, so that a circular disk was placed between two platens and
compressed producing a nearly uniform tensile stress distrib-
ution normal to the loaded (vertical) diametral plane, leading
to the failure of the disk by splitting. The tensile strength was
calculated as σt = (2P/πTD), where P is force, and T and D are
the thickness and diameter of the disk [9]. The microstructure
of the specimens was observed in scanning electron micro-
scope ZEISS (Germany).
3. Results and discussion
3.1. Influence of the monomers on viscosity of the
slurries
As shown in Fig. 3., all of the viscosity curves reveal
the shear-thinning behavior of the suspensions. All of the sus-
pensions present low viscosity, low enough to fill the complex
moulds in details, even despite very high alumina content (over
50vol %). However some differences are visible. The amount
of dispersants were matched individually for the slurries with
different monomers, on the basis of viscosity curves obtained
for the suspensions where 3wt % of monomers were applied.
Both monomers have a positive impact on the viscosity. 5wt %
addition of each monomer decreases the viscosity, that is lower
than the viscosity of the slurry with 3wt % monomer content.
Moreover, the viscosities of the suspensions with HEA are
lower than with Akr-X, even despite the higher solid content.
That was to be expected, because HEA is a low-viscous liquid
whereas Akr-X is a high-viscous resinous substance.
Fig. 3. Viscosity curves for alumina suspensions with HEA and
Akr-X (compositions according to Table 1)
1301
The monomer content in the suspension is a crucial issue,
because after the polymerization, that macromolecular net-
work is responsible for the mechanical strength of the green
elements. Generally, the larger the network is, the stronger the
green bodies, so higher content of the monomer in the sus-
pension leads to higher mechanical strength of the specimens
before the sintering. However, the more organic substances
are added to the suspension, the more difficult it is to burn
them out during sintering, so the easier it is to cause any
defects. Usually the amount of the monomer in the gelcasting
procedure should be not larger that 5wt % and the amount of
3wt % is thought to be the minimum quantity.
3.2. The idle time control
The composition of activator (catalyst) and initiator is
essential additive to start the polymerization in situ in the
ceramic suspension. Their contents also allow to control the
idle time of the process, it means the time interval between
adding the initiator to the suspension and the real start of the
reaction. The creation of the polymer leads to the rapid and
big increase of the viscosity of the slurries. From the tech-
nological point of view, the idle time shouldn’t be then too
quick, as all the operations concerning mixing and filling the
suspension to the mould regard some time. However, the idle
time shouldn’t be too long, because of the simple economi-
cal issues – the whole shaping process should be as fast as
possible. At the same time, the initiator content has a large
influence on the mechanical properties of the green bodies.
The mechanism of polymerization assumes creating free radi-
cals and chain reaction. Too high initiator content leads to the
creation of too many free radicals that start the polymerization
and finally a lot of short polymeric chains are created. On the
other hand, too low initiator content might be not efficient.
The ideal situation is when only few long polymeric chains
are created.
For gelcasting the elements for that research, the poly-
merization was initialized by 0.3 and 0.1 wt % of APS for the
slurries with Akr-X and HEA respectively, having the idle
time around 25 minutes. Table 2 lists the values of the idle
time for different amount of initiator.
TABLE 2
The idle time values controlled by the initiator content
activator initiator idle
monomer content content time
[wt %] [wt %] [min]
2 0.3 25
Akr-X 2 0.5 20
2 0.7 18
2 1.0 14
2 0.1 23
HEA 2 0.3 9
2 0.5 6
2 1.0 4
1302

3.3. Properties of the ceramic bodies

Table 3 shows the results of characterization of the green

and sintered gelcast elements. First of all, it has to be no-
ticed that it was not possible to obtain the specimens from the
suspension with 3wt % of commercial monomer HEA together
with 2wt % cross-linking agent. The exfoliation of the elements
was highly visible, especially while demoulding, the elements
disintegrated or split off into layers while machining, and the
shape of the mould was not reflected at all. Increased amount
of the monomer made it possible to get a desired shape. At
the same time, the specimens obtained by applying new Akr-X
without the external cross-linker for the gelcasting procedure
were easy to demould and machine without any damage.

Fig. 4. Photo of the samples after ‘Brazilian test’ accompanied with

the function of force vs strain registered during the test, a) sample
TABLE 3
with poly(Akr-X), b) sample with poly(HEA)
The properties of the green elements

Relative density Tensile strength

Slurry (denotation)
[% TD] [MPa]
TX3 59.1 ±0.6 4.3 ±0.6
TX5 57.4 ±0.5 5.0 ±0.5
THA3 - -
THA5 60.1 ±0.4 -

Table 3 lists the values of relative density and tensile

strength of the green elements. The densities are compara-
ble, certainly the elements shaped by polymerizing HEA are
more densified due to higher solid content of the suspension.
However, the tensile strength of the green bodies can’t be
compared. The ‘Brasilian test’ used for the calculation of that
value assumes that the tensile force causes the split through the
diameter of the sample parallel to the direction of the force. Fig. 5. SEM microstructure of green (a,b) and sintered (c,d) elements,
That could not be measured for the samples with polymerized obtained by polymerizing in situ 3wt % Akr-X (a,c) and 5wt % Akr-X
HEA, because of the exfoliation. Fig. 3 shows the photographs (b,d)
of the samples after and during the ‘Brazilian test’ accompa-
nied with the functions of force vs strain. It can be noticed,
that even small forces cause exfoliation of the sample contain- TABLE 4
ing polymerized HEA, whereas the samples with poly(Akr-X) The properties of the sintered elements
split properly. Their tensile strengths are then very high, over
Linear shrinkage Relative density
4 MPa, comparing to values reported in other papers. Surely, Slurry (denotation)
[%] [%TD]
the hydrogen bonds formed between the polymer chains are
TX3 17.2 ±0.9 99.2
strong enough to provide a well cross-linked network, without
additional binders. Certainly, the higher polymer content in the TX5 17.8 ±0.8 98.9
sample, the higher mechanical strength of the green bodies,
but the amount of 3wt % of Akr-X is enough to get satisfacto-
ry properties. The homogeneous green microstructure can be
seen at SEM photo of a fracture surface (Fig. 5a-b). More- 4. Conclusions
over, polymeric bridges are visible when 5wt % of monomer is
applied. Following the synthesis route elaborated for acryloyl
Table 4 lists the shrinkage and density values after sinter- derivatives of monosaccharides, similar derivative of xylitol,
ing the bodies with poly(Akr-X) at 1500◦ C for 2h. The den- a polyhydroxyalcohol, was synthesized. This monomer was
sities of the elements are very high, around 99%. It’s clearly applied for a gelcasting of alumina powder. It appears that
visible, that the more organic content in the sample, the more 1-O-acryloylxylitol (Akr-X) is an effective processing agent,
difficult it is to get full densification, because more pores are it does not only play the role of monomer polymerizing in situ
likely to be created, that is also confirmed by the SEM images in the suspension, but also does not need external cross-linker
(Fig. 5c-d). to create a strong macromolecular network. The specimens

obtained by gelcasting a suspension with Akr-X are charac-
terised by very high values of tensile strength of green bod-
ies, as well as very high densities of sintered elements. The
quality of the products is incomparably better than if com-
mercial 2-hydroxyethyl acrylate (HEA) together with MBAM
as cross-linking agent is applied for the same procedure. The
research shows that not only acryloyl derivatives of monosac-
charides, but generally of polyhydroxycompunds are promis-
ing group of substances to use in the process.
Acknowledgements
This work was partially supported by the Ministry of Science
and Higher Education of Poland (Grant No. N N209 19 97 38).
REFERENCES
[1] O.O. O m a t e t e, M.A. J a n n e y, S.D. N u n n, J. Eur. Ce-
ram. Soc.17, 407 (1997).
Received: 20 September 2013.
1303
[2] A.C. Y o u n g, O.O. O m a t e t e, M.A. J a n n e y, P.A.
M e n c h h o f e r, J. Am. Ceram. Soc. 74, 612 (1991).
[3] J. Y a n g, J. Y u, Y. H u a n g, J. Eur. Ceram. Soc. 31, 2569
(2011).
[4] P. B e d n a r e k, Y. S a k k a, M. S z a f r a n, T. M i z e r s -
k i, J. Eur. Ceram. Soc. 30, 1795 (2010).
[5] P. F a l k o w s k i, P. B e d n a r e k, A. D a n e l s k a,
T. M i z e r s k i, M. S z a f r a n, J. Eur. Ceram. Soc. 30, 2805
(2010).
[6] M. S z a f r a n, P. B e d n a r e k, A. S z u d a r s k a, Ad-
vances in Science and Technology 62, 163 (2010).
[7] A. S z u d a r s k a, T. M i z e r s k i, M. S z a f r a n, Archives
of Metallurgy and Materials 56, 1211 (2011).
[8] C. T a l l o n, D. J a c h, R. M o r e n o, M.I. N i e t o, G. R o -
k i c k i, M. S z a f r a n, J. Am. Ceram. Soc. 90, 1386 (2007).
[9] J.R. C a p u a P r o v e t i, G. M i c h o t, Int J Fract. 139, 455
(2006).